African Americans with HIV fare worse than their White counterparts, with large racial/ethnic disparities at all points along the HIV care continuum. Compared to Whites with HIV, African Americans with HIV have lower levels of engagement and retention in care, and are less likely to be on state-of-the art antiretroviral treatment (ART); those on ART are less likely to be adherent at high enough levels for the treatment to be effective. The proposed 5-year project focuses on HIV-related medical mistrust (e.g., misconceptions, conspiracies such as that HIV is a form of genocide) as a barrier to HIV care and treatment behaviors. Specifically, the project will examine how social contexts, specifically, social networks (i.e., the constellation of social relationships surrounding an individual), may influence the spread of HIV-related medical mistrust, and in turn, affect treatment behaviors among African Americans with HIV. A total of 240 African Americans with HIV (120 not engaged in care, 120 on ART) will be followed over 12-months; data will be collected on the flow of HIV-related mistrust beliefs across social networks, as well as HIV care outcomes over time (engagement in care, adherence to ART). The research will conclude with active intervention conceptualization in partnership with community stakeholders in order to identify innovative social network-based solutions for improving treatment behaviors and ultimately, reducing disparities. The specific aims are: (1) To identify social network sources of HIV-related information (i.e., mistrust, misconceptions, conspiracies, and accurate information) among African Americans with HIV; (2) To examine the ways in which social network characteristics related to HIV mistrust are associated with HIV treatment behaviors across the HIV care continuum (engagement in care, antiretroviral treatment use, antiretroviral treatment adherence); and (3) Together with community stakeholders, to use study findings to identify novel intervention solutions to address mistrust and improve HIV treatment behaviors for African Americans with HIV. HIV-related mistrust is hypothesized to be spread through social networks and associated with specific network characteristics; individuals with high HIV-related mistrust are hypothesized to be less likely to be engaged in care and to be adherent to ART; and social network composition (e.g., level of mistrust in network) and structure (e.g., closeness of network members) are predicted to be related to HIV treatment behaviors (e.g., the relationship between mistrust and adherence will be greater for individuals in networks with high levels of mistrust). No research to date has used a social network approach to understanding the flow of HIV-related information and misconceptions among people with HIV.